• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Kangaroo jumps

Francois

Joined
3/30/07
Messages
15
Points
11
Being half aussie, I can't resist to not give this one:
you have a Kangaroo on a 2-D plane. It only makes jumps of length pi and 1, and only
along the x-axis and the y-axis (either jump length along the x and y axes).
Assume the plane large enough (larger than pi x pi ), if you pick a random point (x0,y0)
on the plane, and a random small number epsilon, what is the probability that at some stage
P( dist(Kangaroo,(x0,y0)) < epsilon ?
 
OK, looks like you guys don't like Kangaroos. So, I'll reword the problem. Show that if you play with your money and can either win or lose by $1 or by $pi then, 'given enough time', you can hit any amount (positive or negative)
 
hahah Kangaroos are fine dude ... people here enjoying spring vacation..
 
Spring break was 3 weeks ago! ;-)
would be the first time texas was ahead of NYC for anything ;-)
 
Francois said:
OK, looks like you guys don't like Kangaroos. So, I'll reword the problem. Show that if you play with your money and can either win or lose by $1 or by $pi then, 'given enough time', you can hit any amount (positive or negative)
Click to expand...

Okay, let me give it a start atleast...

Lets break the number line into parts :- the set of integers, and the set of non integers... Now its easy to recognize that the problem is a 2 dimensional random walk... For the set of integers, we know that any level can be hit by the "standard" - in the sense that it can go up or down by 1- random walk which is one of the components of the problem --> so we conclude the set of integers can be hit...

If we show that any number = p*1 + q*pi , p and q being integers (I guess there is nothing sacred about pi here, any other irrational would be fine) then we would be through... Just that am not able to show that -- any hints or am I totally off the track...
 
You're right on actually. The idea is to show that pZ + qZ with p/q irrational is dense in R.
So, it works with other numbers than 1 and pi.
You just need to pick a random element x in R, and build a sequence un in pZ+qZ that converges towards x.
 
Back
Top